www.elsevier.com/locate/scitotenv

Science of the Total Environmen

Review

Spatial and temporal trends of contaminants in terrestrial

biota from the Canadian Arctic

Mary Gamberg a,*, Birgit Braune b, Eric Davey c, Brett Elkin d, Paul F. Hoekstra e,1,

David Kennedy d,2, Colin Macdonald f, Derek Muir g, Amar Nirwal h,

Mark Wayland i, Barbara Zeeb h

aGamberg Consulting, Box 10460, Whitehorse, Canada, YT, Y1A 7A1bCanadian Wildlife Service, Environment Canada, National Wildlife Research Centre, Carleton University,

Raven Road, Ottawa, Canada, ON, K1A 0H3cAthabasca Tribal Council, Environmental Affairs, 9206 McCormick Drive, Fort McMurray, Canada, AB, T9H 1C7

dNorthwest Territories Department of Resources, Wildlife and Economic Development, Yellowknife, Canada, NT X1A 3S8eDepartment of Environmental Biology, University of Guelph, Guelph, Canada, ON, N1G 2W1

fNorthern Environmental Consulting, Pinawa, Canada, MB, R0E 1L0gNational Water Research Institute, Environment Canada, Burlington, Canada, ON, L7R 4A6

hDepartment of Chemistry and Chemical Engineering, Royal Military College of Canada, Box 17000, Stn Forces,

Kingston, Canada, ON, K7K 7B4iCanadian Wildlife Service, Environment Canada, Prairie and Northern Region, 115 Perimeter Road, Saskatoon, Canada, SK, S7N 0X4

Accepted 14 October 2004

Available online 16 August 2005

Abstract

Contaminants in the Canadian Arctic have been studied over the last twelve years under the guidance of the Northern

Contaminants Program. This paper summarizes results from that program from 1998 to 2003 with respect to terrestrial animals

in the Canadian Arctic. The arctic terrestrial environment has few significant contaminant issues, particularly when compared

with freshwater and marine environments. Both current and historical industrial activities in the north may have a continuing

effect on biota in the immediate area, but effects tend to be localized. An investigation of arctic ground squirrels at a site in the

Northwest Territories that had historically received applications of DDT concluded that DDT in arctic ground squirrels livers

was the result of contamination and that this is an indication of the continuing effect of a local point source of DDT. Arsenic

concentrations were higher in berries collected from areas around gold mines in the Northwest Territories than from control

sites, suggesting that gold mining may significantly affect arsenic levels in berries in the Yellowknives Dene traditional territory.

Although moose and caribou from the Canadian Arctic generally carry relatively low contaminant burdens, Yukon moose had

0048-9697/$ - s

doi:10.1016/j.sc

* Correspondin

E-mail addre1 Current addr2 Current addr

Street, Gatineau

t 351–352 (2005) 148–164

ee front matter D 2005 Elsevier B.V. All rights reserved.

itotenv.2004.10.032

g author. Tel.: +1 867 668 7023; fax: +1 867 668 7024.

ss: mary.gamberg@northwestel.net (M. Gamberg).

ess: Golder Associates Ltd., Environmental Sciences Group, Mississauga, Canada, ON, L5N 5Z7.

ess: Environment and Renewable Resources, Department of Indian and Northern Development, Room 640, 10 Wellington

, Canada, QC, K1A 0H4.

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164 149

high renal selenium concentrations, and moose and some woodland caribou from the same area had high renal cadmium levels,

which may put some animals at risk of toxicological effects. Low hepatic copper levels in some caribou herds may indicate a

shortage of copper for metabolic demands, particularly for females. Similarities in patterns of temporal fluctuations in renal

element concentrations for moose and caribou suggest that environmental factors may be a major cause of fluctuations in renal

concentrations of some elements. Concentrations of persistent organochlorines and metals in beaver and muskrat from the

Northwest Territories, and carnivores from across the Canadian Arctic were very low and considered normal for terrestrial

wildlife. Two new classes of persistent fluorinated contaminants, perfluorooctane sulfonate (PFOS) and perfluoroalkyl

carboxylates (PFCAs) were found in arctic carnivores and were most abundant in arctic fox and least abundant in mink.

Although trace element concentrations in king and common eider ducks were low and not of toxicological concern, the number

of nematode parasites in common eiders was positively correlated with total and organic mercury concentrations. Future

research should focus on cadmium in moose and caribou, mercury in caribou, and emerging contaminants, with an effort to

sample moose and caribou annually where possible to explore the role of naturally occurring cycles in apparent temporal trends.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Arctic; Canadian Arctic; Contaminant; Terrestrial ecosystem; Organochlorines; Trace metals; Mercury; Cadmium; Caribou; Moose;

Arctic fox; Wildlife

Contents

. . . . . . 149

. . . . . . 150

. . . . . . 150

. . . . . . 151

. . . . . . 151

. . . . . . 152

. . . . . . 152

. . . . . . 155

. . . . . . 157

. . . . . . 159

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1.1. Pathways of contaminant delivery to the Arctic terrestrial ecosystem . . . . . . . . . . . . .

1.2. Bioaccumulation processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2. Contaminants in Arctic terrestrial biota . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.1. Local sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.2. Vegetation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.3. Moose and Caribou . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.4. Small game . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.5. Arctic carnivores. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.6. Waterfowl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . 161

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

1. Introduction

Contaminants in terrestrial animals in the Canadian

Arctic are of concern from both an animal population

and a human health perspective. Contaminants enter-

ing the environment from local and global sources

may have a significant impact on animal populations

if they cause reproductive impairment, physiological

damage, behavioural modification or death. In addi-

tion, many terrestrial animals in the Arctic are used as

a food source by both Aboriginal and non-Aboriginal

people living in the north. Contaminants accumulated

in these animals may be passed on to people who

consume them, and may, in turn, cause health effects

in those people.

Contaminants in the Canadian Arctic have been

studied over the last twelve years under the guidance

of the Northern Contaminants Program (NCP). The

first six years of research (1991–1997) were summar-

ized in the Contaminants in the Canadian Arctic

Assessment Report (Muir et al., 1997) and Braune

et al. (1999) reviewed trends found from that research

in freshwater and terrestrial ecosystems. Concentra-

tions of persistent organochlorine contaminants (OCs)

were found to be very low in land animals and levels

of radionuclides were not considered likely to pose a

human health concern. The major contaminant con-

cerns for terrestrial animals in the Canadian Arctic

were non-essential elements, specifically cadmium

and mercury, with some relatively high cadmium

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164150

concentrations noted in kidneys and livers of some

woodland caribou from the Yukon. Woodland cari-

bou (Rangifer tarandus) from the western Arctic had

higher cadmium levels than barren-ground caribou

from the eastern Arctic, but no clear trend for mer-

cury was found. Conversely, several OCs increased

in a significant west to east gradient in caribou, while

polychlorinated biphenyls (PCBs) in mink (Mustela

vison) increased in a north to south gradient in the

Northwest Territories (NT). Concerns over contami-

nation at Defense Early Warning (DEW) line sites

were noted. The spatial coverage of terrestrial mam-

mals and birds was considered bquite completeQ andpollutant concentrations in these animals were gen-

erally found to be generally lower than more south-

ern species or those from the marine ecosystem.

Temporal trend data were considered too limited to

be strongly predictive of most OCs and metals

because they were based on only two or three sam-

pling intervals.

Phase II of the NCP guided research on contami-

nants in the Canadian Arctic from 1998 to 2003. It

focussed upon questions about the impacts and risks

to human health that may result from current levels of

contamination in key arctic food species as well as

temporal trends of contaminants of concern in key

arctic indicator species and in air. This paper sum-

marizes the results from that program with respect to

terrestrial animals in the Canadian Arctic, and where

possible, discusses geographical and temporal trends.

1.1. Pathways of contaminant delivery to the Arctic

terrestrial ecosystem

Atmospheric transport, and rivers deliver anthro-

pogenic contaminants to northern terrestrial environ-

ments from more southern and industrialized, urban,

areas. Airborne contaminants are removed from the

atmosphere by gas absorption, precipitation and dry

deposition. Many chlorinated organics are present as

gases even at low temperatures and are absorbed from

the gas phase by water, snow and plant surfaces.

Precipitation scavenging of gas and particles from

the air also deposits particle-associated OCs and

metals in snow and rain. Dry deposition is a third

pathway of input of aerosol-bound contaminants to

terrestrial and aquatic ecosystems (Macdonald et al.,

2000). In the Arctic a significant proportion of the

annual precipitation occurs as snow, and a snowpack

covers the ground for most of the year. As a result,

snow has an important influence on the extent and

timing of contaminant delivery to northern ecosys-

tems (Wania et al., 1998). For the more hydrophilic,

less volatile contaminants, transport via ocean cur-

rents may be more important than the air-borne

route (Li et al., 2002). Water borne contaminants

also enter arctic ecosystems from northward flowing

rivers such as the Athabasca/Peace/Slave River sys-

tem which feeds into Great Slave Lake, the Nelson

River/Lake Winnipeg drainage, and other major rivers

flowing into Hudson Bay.

1.2. Bioaccumulation processes

Hydrophobic organics and heavy metals such as

mercury, cadmium and lead are readily adsorbed by

living and dead organic matter such as particulate

organic carbon, waxy plant surfaces, animal mem-

branes and fats. Once adsorbed, the bioavailability

of these chemicals to terrestrial animals will depend

on the properties of the chemical and on the physical,

chemical and biological environment into which it is

released.

Persistent OCs accumulate in organisms due to a

high affinity for lipids and, most importantly, the

biological inertness of the parent chemical or metabo-

lites. For metals, differences in uptake due to specia-

tion of the element (which may be influenced by water

hardness, salinity, redox conditions in sediment, pH

and temperature), as well as the physiological require-

ments and metabolic rate of the organism affect trans-

fer across biological membranes (Heath, 1987). The

protein metallothionein is important in regulating the

accumulation of some metals in the liver and kidneys

of mammals and fish and elimination of metals into

the bile following uptake.

Many factors such as organic carbon content of

soils and sediments, pH and kinetic limitations

influence the amount of a contaminant that can be

released from food particles in the gut and therefore

affect the environmental bioaccessibility. Despite

being tightly bound to particles, membranes, fat

globules or proteins, most OC and metal contami-

nants of concern in the Arctic have been shown in

laboratory studies with invertebrates, fish, mammals

and birds, to be readily assimilated from the diet

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164 151

(Kelly et al., 2004). Transfer within the food web

through food ingestion is the dominant pathway for

uptake of persistent organic chemicals and heavy

metals in terrestrial food webs. Since the food

requirements of an organism are controlled by meta-

bolic rate and production, metabolic rate is linked to

the rate of uptake of contaminants. These pathways

coupled with the slow rate of excretion and meta-

bolism, lead to biomagnification (as reviewed by

Kelly et al., 2004). Kelly and Gobas (2003) found

that biomagnification factors of persistent OCs in

the terrestrial food web are strongly related to octa-

nol-air partition coefficient (KOA). Substances that

exhibited log KOAs N5 and also exhibit octanol-

water partition coefficients (KOW)N100, such as

hexachlorocyclohexane isomers, showed significant

bioaccumulation in arctic terrestrial food-chains.

This implies that some chemicals considered non-

bioaccumulative under the bioaccumulation defini-

tion (log KOWN5) in the Stockholm POPs protocol

(UNEP 2001) could accumulate in terrestrial food

webs.

2. Contaminants in Arctic terrestrial biota

2.1. Local sources

Local sources of OCs and heavy metals are

considered to be only minor contributors to contam-

ination of the Arctic terrestrial environment when

considered on a continental scale. On a local scale,

however, household heating in settlements, burning

of hydrocarbons for electricity and transport, and

incineration and open burning of garbage, may

contribute significantly to the input of organic pol-

lutants such as PAHs as well as heavy metals such

as mercury. Emissions of polychlorinated dibenzo-p-

dioxins and dibenzofurans (PCDD/Fs) from these

combustion sources may also be important in north-

ern Canada although considered very minor com-

pared to heavily populated regions of North

America (Commoner et al., 2000).

Current and former military bases throughout the

circumpolar Arctic, especially those with older radar

equipment, have been previously identified as

sources of PCBs and heavy metals (de March et

al., 1998; Dietz et al., 1998). Large amounts of

potentially toxic contaminants were stored at DEW

line sites across the Canadian Arctic and no cleanup

was done following the closure of these sites. This

has resulted in a large number of dhazardous waste

sitesT across the north, (Holz et al., 1987) which are

now undergoing remediation in an attempt to miti-

gate their environmental impacts. DEW line sites

have been shown to be significant sources of lead

and PCBs to the surrounding ecosystem, which may

biomagnify within the food web (Braune et al.,

1999). Many of these sites have undergone cleanup

which includes removal of contaminated equipment

and soils. The cleanup of 21 sites in the Canadian

Arctic is scheduled for completion in 2008 (Canada

DND, 2001).

In addition to PCBs, some former military sites

received significant DDT applications. The bioavail-

ability of this localized DDT contamination to the

terrestrial arctic environment was examined in a

study at an abandoned Long Range Aid to Naviga-

tion (LORAN) station located at Kittigazuit, North-

west Territories in the western Canadian Arctic

(Nirwal, 2001). The study site received applications

of DDT between 1948 and 1950. Despite the pas-

sage of time, soil concentrations have remained high,

and the composition of DDT compounds in soil still

resembled the original pesticide formulation. Sam-

ples of soil, sediment, willow (Salix sp.), grass (Ely-

mus sp.), and arctic ground squirrel (Spermophilus

panyi) collected at the LORAN station had higher

concentrations of DDT than those collected from a

nearby reindeer herding camp and background sites.

Hepatic concentrations of total DDT in arctic ground

squirrels at contaminated areas declined to back-

ground levels with increasing distance from contami-

nated areas. Although a significant relationship

between liver size and DDT concentration was

found, estimated contaminant exposures were below

dno-observed effectT levels. The contribution of atmo-

spheric dispersal and transport at Kittigazuit is

believed to be negligible, because an abrupt transition

exists between soil contaminant levels at the sites, and

samples collected immediately off-site. Nirwal (2001)

concluded that the concentration and composition of

DDT in arctic ground squirrels livers were the result

of contamination at the study site and that this is a

clear indication of the continuing effect of a local

point source of DDT.

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164152

2.2. Vegetation

Vegetation is the basis of the terrestrial food web in

the Arctic. Some perennial plants, such as mosses and

lichens, lack root systems and absorb contaminants,

along with their nutrients, from the atmosphere. This

renders them valuable indicators of atmospheric

deposition of contaminants (Thomas et al., 1992).

Other plants absorb their nutrients from the soil, and

in some cases absorb, or even hyperaccumulate, con-

taminants at the same time (Crowder, 1991). These

plants can be used as indicators of local contaminants,

both natural and anthropogenic.

Braune et al. (1999) concluded that although OC

and metal concentrations in arctic plants were gener-

ally low and not of concern, elevated levels of some

contaminants, such as PCBs and lead, were found at

some locations. It was also noted that element con-

centrations in vegetation tend to be extremely variable

depending on plant species, plant tissue and local soil

conditions.

Davey (1999) examined arsenic levels in berries in

the Weledeh Yellowknives Dene traditional area in

response to concerns regarding local gold mining

activity. The study took place in and around Yellow-

knife, NT and included both active and abandoned

gold mines. Analyzed berries included raspberry

(Rubus isaeus), blueberry (Vaccinium ovalifolium),

cranberry (Vaccinium vitis-idaea), rose hip (Rosa aci-

cularus) and gooseberry (Ribes lacustre). Arsenic

concentrations were significantly higher in berries

collected from around gold mines than from control

sites. Most berries harvested at mine sites and some

harvested within the city of Yellowknife were above

the maximum concentration of 0.1 Ag/g recommended

by Health Canada for consumption of fruit juices and

beverages (Health Canada, 1991). The authors con-

cluded that gold mining may significantly affect

arsenic levels in these species of berries in the Yellow-

knives Dene traditional territory.

2.3. Moose and Caribou

Caribou and moose (Alces alces) are a major part

of the social and cultural identity of aboriginal and

northern culture in North America and comprise a

large portion of traditional diets in some areas of

the north. There is continuing concern about the

presence of contaminants in these species, and the

potential effect on human and animal health.

Although organic contaminants such as dichlorodi-

phenyltrichloroethane (DDE), PCBs, PCDD/Fs have

been shown to be low in large terrestrial mammals

(Elkin and Bethke, 1995; Hebert et al., 1996; Tho-

mas et al., 1992), there are concerns that biomagni-

fication factors in the lichen-caribou-wolf food chain

are high (Kelly and Gobas, 2003). Inorganic contami-

nants like cadmium, lead, mercury and radionuclides

have been shown to be elevated under some condi-

tions. The contribution of caribou to total dietary

intake of cadmium, lead and mercury may be relatively

small (b10%) in Inuit communities that also consume

large amounts of marine mammals (Chan et al., 1995),

but moose and caribou liver can comprise up to 90% of

the total cadmium ingested in the Dene/Metis tradi-

tional diet in the Northwest Territories (Berti et al.,

1998). There is also concern that the high metal levels

in organs may lead directly or indirectly to compro-

mised health in some moose and caribou.

A hunter survey program conducted by the Yukon

Contaminants Committee and a body condition study

of the Porcupine caribou herd conducted by Yukon

Environment, Yukon Territorial Government were

continued from 1998–2003, building on annual

moose and caribou data collected from 1994 (Gam-

berg et al., 2005a; Gamberg, 2000). Although samples

were collected from a number of small woodland

caribou herds, for most herds the sample size was

low (b10 over the 5-year period). Only the barren-

ground Porcupine caribou herd had samples sizes N10

annually, and will be discussed here. In the Northwest

Territories and Nunavut, three barren-ground caribou

herds were sampled for the second time, with 4–7

years between sampling (Macdonald et al., 2002).

Kidneys, and in some cases liver tissue, were analyzed

for a suite of elements.

Most elements measured in moose and caribou

tissues were low and did not approach concentrations

that would be considered of toxicological concern

(Tables 1, 2 and 3). Hepatic selenium levels found

in Yukon moose were somewhat higher than those

found in moose from Sweden (Frank et al., 2000) and

Norway (Frøslie et al., 1984) and fell within the

chronic toxicity range for domestic cattle (Puls et

al., 1994). It should be noted that selenium tends to

be elevated in the black shale formations of the Sel-

Table 1

Mean element concentrations and standard deviations (SD) in kid-

neys from Yukon moose and Porcupine caribou collected 1994–

2003 (Agd g�1 wet weight)

Moose (N =481) Porcupine caribou (N =331)

Mean SD Mean SD

Age (years) 4.7 2.8 5.4 2.5

Moisture (%) 79.6 1.8 78.5 1.7

Aluminum 0.34 2.33 0.58 3.99

Arsenic 0.06 0.47 0.09 0.38

Cadmium 27.87 91.55 9.08 27.78

Chromium 0.21 0.72 0.28 1.18

Cobalt 0.09 0.22 0.12 0.29

Copper 3.37 4.80 5.24 3.96

Lead 0.04 1.13 0.09 1.96

Mercury 0.02 0.10 0.41 0.89

Molybdenum 0.27 0.55 0.29 0.57

Selenium 1.02 2.00 1.01 1.64

Zinc 29.39 41.38 25.03 19.77

Data from Gamberg et al. (2005a) and Gamberg unpublished data.

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164 153

wyn Basin, Yukon, and can be found at very high

background levels in some Yukon locations. It is not

surprising, therefore, that some moose from these

regions accumulated relatively high concentrations

in their organs.

Table 2

Geometric means and ranges of elements in kidneys from caribou herds f

Herd Beverly Bluenos

Year 1994 2000 1994

N 11 20 9

Age (years) 6.5 – –

SD 2.4

Moisture (%) 79 79 78

SD 2 1 3

Aluminum 0.7 7.35 0.3

Range 0.36–1.37 2.67–20.3 0.21–

Cadmium 29.3 45.6 30.2

Range 19.3–44.4 24.9–83.4 10.9–

Chromium 0.39 0.22 0.22

Range 0.31–0.49 0.08–0.63 0.17–

Copper 24 21.9 26.5

Range 21.5–26.6 16.7–28.7 21.6–

Lead 0.11 0.55 0.06

Range 0.07–0.19 0.34–0.89 0.04–

Mercury 9.42 6.15 8.27

Range 6.53–13.6 5.64–8.16 6.39–

Zinc 116 121 124

Range 104–129 92.0–159 111–

Values for age and moisture are arithmetic means with standard deviation

Cadmium concentrations in Yukon moose appear

to be high relative to moose from other areas, with the

possible exception of Alaska (Frøslie et al., 1986;

Scanlon et al., 1986; Glooschenko et al., 1988; Brazil

and Ferguson, 1989; Crichton and Pacquet, 2000;

O’Hara et al., 2001). Some of the moose from this

study, particularly those from the southeastern Yukon,

had renal cadmium concentrations that fell within, or

exceeded the threshold range of 80–160 Agd g�1 (wet

weight) at which renal tubule dysfunction has been

shown to occur (Kjellstrom, 1986). Sublethal effects

would be expected at a much lower level (30 Agd g�1

wet weight; Outridge et al., 1994) and would be

expected in 29% of the moose analyzed in this

study. This indicates potential for older moose in

some parts of the Yukon to be at risk due to high

renal cadmium levels.

Analysis of Yukon moose renal element concen-

trations along with the region’s stream sediment

element concentrations suggests that moose renal

cadmium levels (and possibly arsenic and lead

levels) were affected by the underlying geology of

the local environment. The most likely mode of

transfer of these elements from sediment to moose

is via plants growing in the soil and being con-

rom NT and Nunavut (Agd g�1 wet weight)

e Kimmirut

1998 1992 1999

11 10 19

– 4.6 5.4

1.8 2.8

79 80 77

2 3 5

3.92 12 7.92

0.42 3.40–4.54 5.69–25.2 .77–9.27

15.3 26.1 22.5

83.8 8.41–27.8 12.4–54.9 11.7–43.4

0.47 1.55 0.7

0.29 0.43–0.51 0.70–3.43 0.58–0.84

22 28.5 14.7

32.4 14.9–32.3 21.3–38.2 13.3–16.2

0.25 0.4 1.84

0.09 0.14–0.45 0.22–0.73 1.46–2.31

1.92 12.8 3.13

10.7 1.03–3.59 871–18.7 2.28–4.3

105 93 69.2

138 78.2–141 76.6–113 60.7–79.0

(SD). Data from Macdonald et al. (2002).

Table 3

Geometric means (Agd g�1 wet weight) and ranges of elements in livers from caribou herds from NT and Nunavut

Herd Beverly Bluenose Kimmirut

Year 1994 2000 1994 1998 1992 1999

N 10 20 10 12 10 19

Age (years) 6.5 – – – 4.6 5.4

SD 2.4 1.8 2.8

Moisture (%) 70 71 70 70 73 69

SD 2 1 2 1 5 1

Aluminum 0.89 5.61 0.37 2.73 7.25 7.04

Range 0.33–2.40 2.01–15.7 0.19–0.73 2.31–3.22 2.23–23.6 5.69–8.72

Cadmium 3.32 5.2 4.77 3.12 3.83 3.47

Range 2.56–4.3 3.35–8.06 2.24–10.2 1.99–4.88 2.16–6.80 2.23–5.39

Chromium b0.40 0.19 b0.20 0.33 0.09 b0.01

Range 0.08–0.47 0.32–0.34 0.01–1.15

Copper 60.1 18.5 27.5 130 100 103

Range 28.1–129 10.1–34.1 18.0–42.1 85.3–199 40–250 73.3–143

Lead 0.11 1.08 0.09 0.39 2.82 7.76

Range 0.09–0.15 0.60–19.5 0.04–0.19 0.27–0.55 1.54–5.19 5.32–11.3

Mercury 1.17 0.8 1.48 0.46 2.04 0.75

Range 0.85–1.62 0.46–1.38 0.95–2.30 0.31–0.69 1.12–3.73 0.54–1.06

Zinc 82.9 87.7 79.5 105 71.3 62.6

Range 60.5–114 68.3–112 66.5–95.2 85.9–129 60.1–84.5 55.8–70.2

Values for age and moisture are arithmetic means with standard deviation (SD). Data from Macdonald et al. (2002).

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164154

sumed by moose. Willows (Salix sp.) are a preferred

food species for Yukon moose (Risenhoover, 1989),

and have been shown to be hyperaccumulators of

cadmium (Vandecasteele et al., 2002). Other plants

have been shown to accumulate arsenic (Ma et al.,

2001) and lead (Lasat, 2002). Geographical differ-

ences seen in sediment and moose renal cadmium

and the absence of a latitudinal gradient suggest a

local rather than a global source of cadmium.

Although the results do not rule out an anthropo-

genic source of cadmium, they do indicate that there

are natural sources of high cadmium levels in the

Yukon that likely contribute significantly to the cad-

mium body burden of local wildlife (Gamberg et al.,

2005a).

Although barren-ground caribou from Yukon, NT

and Nunavut had lower renal cadmium levels than

Yukon moose, some individuals and the average renal

cadmium concentration for the Beverly herd exceeded

the threshold level of 30 Agd g�1 wet weight at which

sublethal effects might be expected (Outridge et al.,

1994)(Tables 1 and 2). Macdonald et al. (2002)

reported that relatively low hepatic copper levels in

some herds may indicate a shortage of copper for

metabolic demands, particularly for females.

Because renal cadmium levels often increase

with age of the animal (Friberg et al., 1992), it is

important to consider age as a co-factor when com-

paring cadmium levels among herds, or with time.

This is likely the reason for the high standard

deviation seen around mean cadmium concentra-

tions in Porcupine caribou (Table 1). Macdonald

et al. (2002) found a positive relationship between

caribou age and renal mercury concentration. Unfor-

tunately, age was unavailable for several of the NT

collections. Using renal element/age as a simple

correction for mean renal cadmium and mercury

concentrations, there was some evidence of a geo-

graphical trend among barren-ground caribou herds,

with higher levels of cadmium being found in the

eastern Arctic (Fig. 1). This may reflect an east/west

gradient in atmospheric deposition of cadmium,

which would be absorbed by lichens, a preferred

forage for barren-ground caribou (Kelsall, 1968).

This trend differs from the east/west gradient in

renal cadmium in arctic caribou described in Braune

et al. (1999) in which western herds had higher

concentrations. This trend was largely determined

by three woodland caribou herds from the Yukon,

which had very high cadmium levels. Given that

KimmirutCaribou

1992 1999

PorcupineCaribou

1992 1999

1994

Cadmium/age

Mercury/age0

2

4

6

0

2

4

6

Con

cent

ratio

n(µ

g.g-1

wet

wt)

BeverlyCaribou

Fig. 1. Mean kidney element concentrations corrected for age in

barren-ground caribou from the Canadian Arctic. Data from Mac-

donald et al. (2002) and Gamberg unpublished data.

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164 155

woodland caribou forage on browse (including cad-

mium-hyperaccumulating willows) over the winter,

while the barren-ground subspecies feed exclusively

on lichens during that time, it is not surprising that

the woodland caribou have higher cadmium concen-

trations. As a result, it is reasonable to consider the

two subspecies separately when evaluating the

potential spatial and (or) temporal trends of cad-

mium and other contaminants in this species. The

potential for a geographical trend in renal mercury

in barren-ground caribou is less clear due to an

apparent decline over time in some herds.

Temporal trend analysis of the Yukon data for

moose and caribou indicated that although aluminum

and cadmium did not change significantly from

1994–2003, arsenic, copper, lead and mercury

showed a decline, while selenium and zinc showed

an increase over the same time period. Although

these trends were statistically significant ( p b0.05),

none of them exhibited a clear and steady change

over time, and there was considerable inter-year

variation (Fig. 2). Because these data encompass

only ten years, it is unclear whether these are true

trends or whether they are part of inter-year sample

variation, or naturally occurring cycles. Macdonald

et al. (2002) reported a decrease in renal mercury

over time in the Kimmirut caribou. As this data was

from only two years, it is not possible to determine

whether this is a real trend or part of a naturally

occurring cycle. Ongoing monitoring should clarify

these potential trends.

Macdonald et al. (2002) also reported seasonal

differences in renal element concentrations in caribou,

with mercury and cadmium being higher in spring-

collected samples, consistent with lower organ

weights in spring than fall (Gerhart et al., 1996).

Copper concentrations were significantly higher in

fall than spring, probably due to an accumulation of

the micronutrient over the summer.

Of particular interest is the similarity between

moose and caribou in the yearly fluctuations of

many of the elements studied (Fig. 2). Temporal

fluctuations seen in some elements such as aluminum,

arsenic and zinc were very similar for both species,

while others, such as mercury and cadmium were very

different. Selenium seemed to show a lag effect with

renal concentrations in moose following those in car-

ibou by one year. The similarities in patterns between

the two species suggest that environmental factors

may be a major cause of fluctuations in renal concen-

trations of some elements. Cadmium and mercury

showed different fluctuation patterns, and also showed

significant differences in absolute renal concentrations

between the species, moose having higher cadmium

levels, and caribou having higher mercury concentra-

tions. This may be explained by differences in diet.

Hyperaccumulators of cadmium, such as willow (Van-

decasteele et al., 2002), are a preferred food for moose

(Risenhoover, 1989) while barren-ground caribou

feed more on lichens (Kelsall, 1968). These data

suggest that lichens may be lower in cadmium, but

higher in mercury than willows. Environmental fac-

tors may impact more heavily on renal element con-

centrations when those concentrations are higher.

2.4. Small game

Beaver (Castor canadensis) and muskrat (Ondatra

zibethicus) meat is an important part of the traditional

diet of the Dene and Metis in the NT. In 1998 and

2001, local trappers collected beaver and muskrat

tissue samples from the southern end (Slave River

Delta) and northern end (Mackenzie River Delta) of

the Mackenzie River watershed (Kennedy, 1999;

Snowshoe, 2003). Beaver and muskrat muscle

(Slave River Delta only) and liver samples were

pooled by species, sex and location for OC analysis.

Overall, total PCBs, DDT and chlordane levels were

low and below available guideline levels, while other

Fig. 2. Temporal trends in mean renal element concentrations (Agd g�1 wet weight) in Yukon moose and Porcupine caribou collected 1994–

2003. Data from Gamberg et al. (2005a) and Gamberg unpublished data.

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164156

OCs were below method detection limits. Levels of

metals in beaver and muskrat muscle were very low

and considered normal for terrestrial wildlife. Average

levels of cadmium in beaver liver and kidney at the

Mackenzie River delta were 10 Agd g�1 (dry weight)

and 55 Agd g�1 (dry weight), respectively while aver-

age levels of cadmium in beaver liver at the Slave

River Delta were 6.6 Agd g�1 (dry weight). There

were no significant differences ( p=0.324) in liver

cadmium concentrations among beaver sampled

Table 4

Mean and (range) of total PFCAs (P

PFCAs) and total PFOS

equivalents (P

PFOS) for arctic carnivores (ngd g�1 wet weight

Species Site and yearP

PFCAaP

PFOSb

Mink Watson Lake,

YT 2001

24

(3–58)

10

(1–22)

Arctic

fox

Arviat,

NU 2001

53

(5–227)

269

(6–1510)

Polar

bear

Sanikiluaq,

NU 2002

325

(214–420)

3112

(1700–4000

Data from Martin et al. (2004).a Sum of PFCAs, including perfluorooctanoate (PFOA), perfluoro

nonanoic acid (PFNA), perfluorodecanoic acid (PFDA), perfluorounde

canoate (PFUnA), perfluorododecanoate (PFDoA), Perfluorotridecanoate

(PFTrA), perfluorotetradecanoate (PFTA), and perfluoropentadecanoic

acid (PFPA).b Sum of PFOS, including perfluorooctane sulfonate (PFOS) and

heptadecafluorooctane sulfonamide (FOSA).

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164 157

from the Mackenzie River delta, the Slave River delta

(Kennedy, 1999) and the Yukon (Gamberg, 2000).

Cadmium concentrations in beaver livers and kidneys

from this study likely reflect natural background

levels and are consistent with other terrestrial wildlife

levels.

2.5. Arctic carnivores

Little work has been done to determine contami-

nant levels in terrestrial carnivores. These species prey

upon lower trophic level species, and therefore, would

be exposed to elevated concentrations of some con-

taminants due to food web magnification. Wolves

(Canis lupus), arctic fox (Alopex lagopus), wolverine

(Gulo gulo) and mink are all circumpolar species that

are widely distributed across the Canadian Arctic.

They are economically important because of their

valuable pelts, and the fox and wolverine are used

as a food source in some communities (Pasitschniak-

Arts and Lariviere, 1995). There is a paucity of infor-

mation on the potential toxicological impacts of envir-

onmentally relevant concentrations of contaminants to

arctic carnivores. One study on NT mink found little

or no effect on reproduction or population health as a

result of contaminants (Poole et al., 1995).

Samples from wolves (Gamberg and Braune,

1999), arctic fox and wolverine (Hoekstra et al.,

2003a, b ,c) and mink (Gamberg et al., 2005b; Martin

et al., 2004) were collected from the Canadian Arctic,

and arctic fox were also collected from Barrow,

Alaska. Their tissues were analyzed for a variety of

elements and organic contaminants.

Recently two classes of persistent fluorinated con-

taminants, perfluorooctane sulfonate (PFOS) and per-

fluoroalkyl carboxylates (PFCAs) were discovered in

wildlife from various urban and remote locations

(Giesy and Kannan, 2001). A preliminary assessment

of these contaminants in the Canadian Arctic

included mink from the Yukon and arctic fox from

Arviat (Martin et al., 2004). Concentrations of PFOS

and PFCAs were highest in polar bears (Ursus mar-

itimus), followed by arctic fox and lowest in mink

(Table 4). In both mink and arctic fox, total PFOS

levels were comparable to total PCB concentrations.

Perfluorononanoate (PFNA) concentrations exceeded

PFOS concentrations in mink, indicating that PFNA

and other PFCAs should be considered in future risk

)

)

-

-

assessments. Little is currently known about the

toxicity of these contaminants and their effects on

wildlife.

Total PCBs and chlordanes were the most abundant

of the OCs measured in arctic fox muscle and liver,

and wolverine liver, while total PCBs and chloroben-

zenes predominated in wolves. OCs were not mea-

sured in the mink study. The major individual PCB

congeners found in all species and tissues measured

were PCB-153 and PCB-180. Hepatic concentrations

of chlordanes in wolverine were generally lower than

in arctic fox, but levels of total PCBs were similar in

both species (Fig. 3). All OCs were lower in wolves.

These results, along with the overall profile of OCs in

wolverine and arctic fox livers, suggest similarities

between these two species in their dietary exposure.

Both fox and wolverine study populations were near

coastal areas and therefore their diets likely included

marine biota. The low levels of OCs in wolf liver

probably reflect the very low levels of these contami-

nants measured in Yukon moose and caribou (Gam-

berg, 2000; Gamberg et al., 2005a), which make up

the bulk of the wolf diet in the Yukon (Hayes, 1995).

The pattern of organochlorine bioaccumulation in

these carnivores is similar to other top predators in

the arctic marine environment and suggests that these

species have a similar ability to metabolize some OCs

as polar bears.

While total PCBs in wolves, arctic fox and wol-

verine from these studies did not exceed concentra-

tions associated with reproductive impairment in mink

0

100

200

Ulukhaqtuuq(Arctic Fox)

Kugluktuk(Wolverine)

Barrow(Arctic Fox)

Yukon(Wolf)

∑ Chlordanesa

∑ PCBsbCon

cent

ratio

n(n

g.g-1

wet

wt)

Fig. 3. Total concentrations of total liver chlordanes (P

Chlordanes)

and PCBs (P

PCBs) in arctic carnivores from the Canadian Arctic.

(Data from Hoekstra et al., 2003a,c). aSum of oxy-, cis- and trans-

chlordane, cis- and trans-nonachlor, and cis-heptachlor epoxide,

and heptachlor (Arctic fox and wolverine only). bSum of PCB-8/5,

19, 18, 15/17, 24/27, 16, 32, 26, 25, 28, 31, 33, 22, 45, 46, 52, 49,

47/48, 44, 42, 64, 41/71, 40, 74, 76/70, 95/66, 91, 56/60, /84, 101,

99, 83, 97, 87, 85, 136, 110, 151, 135, 144, 107/149/118, 114, 153,

105, 141/179, 137, 176/130, 163, 138, 158, 129/178, 175, 187, 182,

183, 128, 185, 174, 177, 171, 156, 172, 197, 180, 193, 191, 199,

170/190, 198, 201, 196/203, 189, 206, 195, 207, 194, 205, 208, and

209. In wolves, also PCB-89, 90, 131,134, 164, 146 and 157. In

arctic fox and wolverine, also PCB-4/10, 7/9, 6, 12/13, 54/29, 50,

21/53, 51, 43, 59, 100, 63, 98, 55, 92, 119, 81, 82, 147, 133, 143,

141, 145, 132, 167 and 202/173.

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164158

(Giesy et al., 1994), further research is recommended

to evaluate the potential impact of exposure to other

OCs, particularly chlordane and its potentially toxic

metabolites, to the overall health of these species in

this region (Hoekstra et al., 2003a, c).

When comparing mercury and cadmium concen-

trations among terrestrial species or locations, it is

important to consider the effect of age, since both

metals have been shown to increase with age, cad-

mium more commonly (Friberg et al., 1992), but

mercury in at least some species (Macdonald et al.,

2002). A simple correction was performed (element

concentration/age) to present comparable cadmium

and mercury values (Fig. 4). Hepatic and renal cad-

mium concentrations were lower in wolverines and

mink than in arctic fox, but these all fell within the

wide range found for wolves across the Arctic (Fig.

4). Cadmium concentrations in wolf liver and kidney

were somewhat higher in Yukon wolves than those

from the NT and Nunavut. This probably reflects the

high cadmium concentrations found in livers and

kidneys of moose and some caribou herds in the

Yukon, as compared to those from the NT (Macdo-

nald et al., 2002; Gamberg, 2000; Gamberg et al.,

2005a). The relatively high concentrations of cad-

mium found in arctic fox liver are likely a reflection

of dietary differences among species, and perhaps

location. Since cadmium levels tend to be higher in

marine biota (Nilsson and Huntington, 2002), this

suggests that marine scavenging may constitute a

greater proportion of the diet in arctic fox than in

mink or wolverine, and that fox from Arviat may

include more terrestrial prey in their diet than those

from Ulukhaqtuuq.

Total mercury concentrations in wolverine and

wolf liver were lower than those found in arctic fox

and mink (Fig. 4). Again, this likely reflects dietary

differences among the species. Unlike wolves and

wolverine, arctic fox and mink include fish in their

diet (Fay and Stephenson, 1989), which can have high

concentrations of mercury (Braune et al., 1999), con-

tributing significantly to the mercury body burden of

those consuming them. Additionally, if arctic fox do

consume a greater proportion of their diet from the

marine ecosystem than the other two species, higher

mercury body burdens would be expected, since mar-

ine biota also tend to have high levels of mercury

(Nilsson and Huntington, 2002). This is consistent

with arctic fox from Arviat having lower cadmium

and mercury levels than those from Ulukhaqtuuq,

perhaps as a result of including more terrestrial prey

in their diet.

An inverse relationship between total mercury and

% MeHg (the proportion of total mercury present as

methylmercury) has been explained in a variety of

animals to be a threshold relationship where mercury

consumed as methylmercury remains in that form

until a concentration of about 10 Agd g�1 (wet weight)

is reached and a demethylation mechanism is acti-

vated (Weiner et al., 2002). Wagemann (1997)

hypothesized that demethylation of methylmercury

leads to the formation of mercuric selenide, resulting

in a positive relationship between total mercury and

selenium concentrations. Data from Yukon mink

reflected both of these trends, with an inverse relation-

ship between total mercury and % MeHg (Fig. 5), and

0

1

0

1

BathurstbYukona

VictoriaIslandb

Cadmium/Age

Mercury/Age

Wolves

0

0.5

1

1.5

0

0.5

1

1.5Arviatd

Arctic fox

Ulukhaqtuuqd

Arctic Fox

Kugluktukd

Wolverine

Cadmium/Age

Mercury/Age

Co

nce

ntr

atio

n

Co

nce

ntr

atio

n

0.5

1.5

Yukonc

Mink

Fig. 4. Mean concentrations of liver cadmium/age (Agd g�1 dry weight/year) and mercury/age (Agd g�1 wet weight/year) in carnivores from the

Canadian Arctic. aData from Gamberg and Braune (1999). bData from Brett Elkin, unpublished data. cData calculated from Gamberg et al.

(2005b). Because liver cadmium levels were unavailable for mink, a correction factor of 5.8 (calculated from arctic fox data) was used to

estimate liver cadmium from kidney cadmium. dData from Hoekstra et al. (2003b).

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164 159

a positive relationship between total mercury and

selenium. The former, however, did not show the

threshold effect that has been demonstrated in other

species. In fact, total mercury levels found were con-

sistently less than half that threshold concentration.

Both relationships were seen most strongly in mink

liver, less so in kidneys and not at all in brains where

most of the mercury was maintained in the methyl

form.

Hoekstra et al. (2003b) suggested that since hepatic

total mercury concentrations in arctic fox from the

current study were not significantly different from

Fig. 5. Total and % methyl mercury in Yukon mink liver. Data from

Gamberg (2005b).

specimens collected in 1973 (Smith and Armstrong,

1975), mercury concentrations have not changed dra-

matically in that population over the past 30 years.

Caution should be used, however, when drawing con-

clusions from a comparison of data from only two

years. Natural cycles or inter-year variation in mer-

cury concentrations, and the age of animals in each

collection may affect results.

Total mercury and cadmium concentrations in

wolves, arctic fox, wolverine and mink tissues from

these studies were well below concentrations asso-

ciated with mercury or cadmium intoxication

(Thompson et al., 1996; Scheuhammer, 1991), and

should be considered baseline levels.

2.6. Waterfowl

Migratory waterfowl not only accumulate contami-

nants from arctic ecosystems, but also from their more

temperate and industrialized wintering grounds.

Braune et al. (1999) found the levels of OCs and

metals in the breast muscle of waterfowl to be gen-

erally quite low. They found that molluscivores and

piscivores from the eastern Arctic tended to have

higher levels of organic contaminants than those

from western locations, but did not find a similar

trend in metal concentrations. Buckman et al. (2004)

concluded that trophic level, migration, scavenging

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164160

and biotransformation all play important roles in

determining concentrations of OCs in arctic seabirds.

Contaminants are believed to be one of several risk

factors that may be contributing to the precipitous

decline of eider ducks (Somateria sp.) in recent

years (Suydam et al., 2000). Cadmium, selenium

and, to a lesser degree, mercury have been found at

elevated concentrations in eiders (Dietz et al., 1996;

Franson et al., 2000). Concentrations of selected trace

elements were determined in livers or kidneys of king

eiders (S. spectabilis) and common eiders (S. mollis-

sima), at three locations in the Canadian Arctic (Way-

land et al., 2001).

Renal and hepatic cadmium concentrations did not

differ between species in the western Arctic, but were

higher in king eiders than in common eiders in the

eastern Arctic where the highest levels recorded in

eider ducks were found (Fig. 6). This is consistent

Belcher's

East Bay

Common eiderKing eiderC

once

ntra

tion

(µg.

g-1 w

et w

t)

Hepatic Mercury

Ulukhaqtuuq(Holman)

0

0.5

1

Belcher’s

East Bay

Ulukhaqtuuq(Holman)

Renal Cadmium

0

25

50

Common eiderKing eiderCon

cent

ratio

n(µ

g.g-1

wet

wt)

Fig. 6. Mean concentrations of liver mercury and selenium, and of kidn

Canadian Arctic. (Data from Wayland et al., 2001).

with reports of high cadmium levels in marine animals

from the eastern Canadian Arctic, which have been

attributed to elevated levels of natural cadmium in the

region’s bedrock (Muir et al., 1997). However, the

relatively low concentrations of cadmium in common

eiders from East Bay contrasted with results for mar-

ine animals. Other factors, such as age of the birds,

may have influenced results. The ratio of cadmium in

liver to that in kidney averaged 0.23 and ranged from

0.09–0.61, a range that is indicative of chronic expo-

sure to low levels of cadmium (Scheuhammer, 1987).

Hepatic mercury and zinc were higher in king

eiders than in common eiders. Foraging habitat and

dietary segregation between the two species may

account, at least partially, for the observed differences

in their trace element concentrations. Both species

feed heavily on mussels. However, whereas common

eiders are mussel specialists, king eiders consume a

Belcher’s

East Bay

Common eiderKing eiderC

once

ntra

tion

(µg.

g-1 w

et w

t)

Hepatic Selenium

0

5

10

15

Ulukhaqtuuq(Holman)

ey cadmium in king and common eiders at three locations in the

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164 161

more varied diet that includes not only mussels but

also echinoderms and other benthic invertebrates

(Bustnes and Erikstad, 1988; Frimer, 1997). In addi-

tion, differential exposure to mercury on their respec-

tive wintering grounds may affect total mercury body

burdens.

Selenium concentrations were significantly higher

in livers of eiders from Ulukhaqtaaq (Holman) than in

those of eiders from East Bay or the Belcher Islands

(Fig. 6), consistent with spatial trends in marine mam-

mals (Muir et al., 1997). It is possible that exposure of

eiders to high levels of selenium during their pro-

longed period of residency in the Bering Sea may

have some residual effect on tissue selenium concen-

trations during spring migration through Ulukhaqtaaq.

Trace element concentrations in these two duck

species were below published toxicity thresholds

and there was no histopathological evidence of kidney

or liver lesions that are typical of trace metal poison-

ing. However, in common eiders, the number of

nematode parasites was positively correlated with

total and organic mercury. This study found no evi-

dence to support the hypothesis that trace metal expo-

sure may be contributing to adverse effects on the

health of individuals of these species.

3. Summary

An investigation of arctic ground squirrels at a site

that had historically received applications of DDT

concluded that the concentration and composition of

DDT in arctic ground squirrels livers were the result

of contamination at the study site and that this is an

indication of the continuing effect of a local point

source of DDT. However, estimated contaminant

exposures were below dno-observed effectT levels.

Arsenic concentrations were higher in berries col-

lected from areas around gold mines in NT than from

control sites and the authors concluded that gold

mining may significantly affect arsenic levels in ber-

ries in the Yellowknives Dene traditional territory.

In general, moose and caribou from the Canadian

Arctic carry very low contaminant burdens. Moose

from the Yukon tend to have high renal selenium

concentrations, and moose and some woodland cari-

bou from the same area tend to have high renal

cadmium levels. Although these high levels appear

to be naturally occurring, coming from the region’s

naturally high mineral content, they may put some

animals at risk of toxicological effects. Low hepatic

copper levels in some caribou herds may indicate a

shortage of copper for metabolic demands, particu-

larly for females. There was some evidence of a

geographical trend among barren-ground caribou

herds, with higher levels of renal cadmium being

found in the eastern Arctic, but no clear trend was

seen in renal mercury. Temporal trend analysis of the

Yukon data for moose and caribou indicated that

although aluminum and cadmium did not change

significantly from 1994–2003, arsenic, copper, lead

and mercury show a decline, while selenium and zinc

show an increase over the same time period. One

barren-ground caribou herd from Nunavut also

showed a decline in mercury between two sampling

years seven years apart. It is unclear whether these are

true trends or whether they are part of natural inter-

year variation, or naturally occurring cycles in ele-

ment concentrations. Similarities in patterns of tem-

poral fluctuations in renal element concentrations for

Yukon moose and caribou suggest that environmental

factors may be a major cause of fluctuations in renal

concentrations of some elements. Ongoing monitoring

should clarify these potential temporal trends, and

should be carried out annually whenever possible to

explore the possibility of naturally occurring cycles

and avoid erroneous conclusions based on data from a

small number of widely spaced sample years.

Concentrations of OCs and metals in beaver and

muskrat from NT were very low and considered nor-

mal for terrestrial wildlife. No geographical trend was

seen in liver cadmium in beavers from the Mackenzie

River delta, the Slave River delta and the Yukon.

Cadmium concentrations in beaver livers and kidneys

reflected natural background levels and were consis-

tent with other terrestrial wildlife levels.

Overall, concentrations of organic and inorganic

contaminants in arctic carnivores were very low and

not of toxicological concern. Two new classes of

persistent fluorinated contaminants, PFOS and

PFCAs were found to be most abundant in arctic

fox and least abundant in mink. Since little is cur-

rently known about the toxicity of these contaminants

and their effects on wildlife, future studies should

continue to monitor PFCAs and PFOS-related con-

taminants and to explore the absolute and relative

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164162

toxicity of these chemicals in wildlife species. A

geographical trend was seen in renal cadmium con-

centrations in arctic wolves, with higher levels in the

west. This likely reflects the higher cadmium in pre-

ferred wolf prey species (moose and caribou) in the

western Arctic as compared with the east. No geogra-

phical trend was apparent for mercury in arctic

wolves. Mercury and cadmium concentrations in

wolves, arctic fox, wolverine and mink tissues were

well below concentrations associated with mercury or

cadmium intoxication and should be considered base-

line levels.

Trace element concentrations in king and common

eider ducks were below published toxicity thresholds

and there was no histopathological evidence of kidney

or liver lesions that are typical of trace metal poison-

ing. However, in common eiders, the number of

nematode parasites was positively correlated with

total and organic mercury concentrations. Although

the ratio of cadmium in liver to that in kidney fell

within a range that is indicative of chronic exposure to

low levels of cadmium, no evidence was found to

support the hypothesis that trace metal exposure may

be contributing to adverse effects on the health of

individuals of these species.

The Canadian Arctic terrestrial environment has

few significant contaminant issues that can be attrib-

uted to long-range transport and deposition of POPs

and heavy metals, particularly when compared with

the freshwater and marine environments. Arctic

small game, such as beavers and muskrat, waterfowl

and arctic carnivores have consistently low body

burdens of organic and inorganic contaminants.

Both current and historical industrial activities in

the north may have a continuing effect on biota in

the immediate area, but effects tend to be localized.

Current significant concerns include cadmium in

moose and caribou, mercury in caribou, and emer-

ging contaminants. Future research should focus on

these issues, with an effort to sample moose and

caribou annually where possible to explore the role

of naturally occurring cycles in apparent temporal

trends. Terrestrial food webs are vulnerable to bio-

magnification of persistent organic contaminants that

have high log KOAs (Kelly and Gobas, 2003; Kelly

et al., 2004). Thus future studies on emerging con-

taminants should consider a wider range of chemi-

cals in terrestrial carnivores than has been

determined to date. The recent detection of PFCAs

and PFOS-related contaminants in mink and arctic

fox illustrate the need for additional monitoring of

new contaminants. There is also a need to explore

the absolute and relative toxicity of these new che-

micals in wildlife species.

Acknowledgements

The authors would like to gratefully acknowledge

the Northern Contaminants Program for financial

support of contaminant research in the Canadian

Arctic, and for providing the structure and guidance

under which that research was conducted. We would

also like to thank the Contaminant Committees in

the Yukon, NT and Nunavut and the Aboriginal

organizations that have supported this research, par-

ticularly the Council of Yukon First Nations and the

Dene Nation. Special thanks go to the northern

hunters and trappers who provided many of the

samples for this research. Without their help and

active participation, much of this research would

have been impossible.

References

Berti PR, Receveur O, Chan HM, Kuhnlein HV. Dietary exposure to

chemical contaminants from traditional food among adult Dene/

Metis in the western Northwest Territories, Canada. Environ Res

1998;76:131–42.

Braune B, Muir D, DeMarche B, Gamberg M, Poole K, Currie R,

et al. Spatial and temporal trends of contaminants in Canadian

Arctic freshwater and terrestrial ecosystems: a review. Sci

Total Environ 1999;230:145–207.

Brazil J, Ferguson S. Cadmium concentrations in Newfoundland

moose. Alces 1989;25:52–7.

Buckman AH, Norstrom RJ, Hobson KA, Karnovskye NJ, Duffe J,

Fisk AT. Organochlorine contaminants in seven species of Arc-

tic seabirds from northern Baffin Bay. Environ Pollut 2004;128:

327–38.

Bustnes JO, Erikstad KE. The diet of sympatric wintering popu-

lations of common eider, Somateria mollissima and king

eider, S spectabilis in northern Norway. Ornis Fenn 1988;

65:163–8.

Canada DND (Department of National Defence). Website;

2001. Available at: http://www.dnd.ca/eng/archive/2001/aug01/

31distant_b_e.htm.

Chan HM, Kim C, Khoday K, Receveur O, Kuhnlein HV. Assess-

ment of dietary exposure to trace metals in Baffin island food.

Environ Health Perspect 1995;103:740–6.

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164 163

Commoner B, Barlett PM, Eisl H, Couchot K. Long range air trans-

port of dioxin from north American sources to ecologically

vulnerable receptors in Nunavut, Arctic Canada. Montreal, QC,

Canada7 North American Commission for Environmental Coop-

eration; 2000.

Crichton V, Pacquet PC. Cadmium in Manitoba’s wildlife. Alces

2000;36:205–16.

Crowder A. Acidification, metals and macrophytes. Environ Pollut

1991;71:171–204.

Davey E. Arsenic levels in berries and soils from the Yellowknife

denes first nation traditional territory. In: Kalhok S, editor.

Synopsis of research conducted under the 1998–99 northern

contaminants program. Ottawa7 Indian and Northern Affairs

Canada; 1999. p. 81–5.

de March BGE, de Wit CA, Muir DCG, Braune BM, Gregor DJ,

Norstrom RJ, et al. Persistent organic pollutants. AMAP

assessment report Arctic pollution issues, Chapter 6. Oslo,

Norway7 Arctic Monitoring and Assessment Programme;

1998. p. 183–372.

Dietz R, Riget F, Johansen P. Lead, cadmium, mercury and sele-

nium in Greenland marine animals. Sci Total Environ

1996;186:67–93.

Dietz R, Pacyna J, Thomas DJ. Heavy metals. AMAP asssessment

report: arctic pollution issues. Oslo, Norway7 Arctic Monitoring

and Assessment Program; 1998. p. 373–524.

Elkin BT, Bethke RW. Environmental contaminants in caribou in

the northwest territories. Sci Total Environ 1995;160/161:

307–22.

Fay FH, Stephenson RO. Annual, seasonal, and habitat-related

variation in feeding habits of the arctic fox (Alopex lagopus)

on St Lawrence Island, Bering Sea. Can J Zool 1989;67:

1986–94.

Frank A, Danielsson R, Jones B. The dmysteriousT disease in

Swedish moose: concentrations of trace elements in liver and

kidneys and clinical chemistry: comparison with experimental

molybdenosis and copper deficiency in the goat. Sci Total

Environ 2000;249:107–22.

Franson JC, Hollmen T, Poppenga RH, Hario M, Kilpi M. Metals

and trace elements in tissues of common eiders (Somateria

mollissima) from the Finnish archipelago. Ornis Fenn 2000;

77:57–63.

Friberg L, Elinder CG, Kjellstrom T. Cadmium. Environmental

health criteria 134. Geneva7 World Health Organization; 1992.

280 pp.

Frimer O. Diet of moulting king eiders Somateria spectabilis

at Disko Island, West Greenland. Ornis Fenn 1997;74:

187–94.

Frøslie A, Norheim G, Rambaek J, Steinnes E. Levels of trace

elements in liver from Norwegian moose, reindeer and red

deer in relation to atmospheric deposition. Acta Vet Scand

1984;25:333–45.

Frøslie A, Haugen A, Holt G, Norheim G. Levels of cadmium in

liver and kidneys from Norwegian cervides. Bull Environ Con-

tam Toxicol 1986;37:453–60.

Gamberg M, Contaminants in Yukon Country Foods. Unpublished

report prepared for the Department of Indian and Northern

Affairs, Whitehorse, Yukon; 2000. 95 pp.

Gamberg M, Braune BM. Contaminant residue levels in arctic

wolves (Canis lupus) from the Yukon Territory, Canada. Sci

Total Environ 1999;243/244:329–38.

Gamberg M, Palmer M, Roach P. Temporal and geographic trends

in trace element concentrations in moose from Yukon, Canada.

Sci Total Environ 2005a;351–352:530–8 [this issue].

Gamberg M, Boila G, Stern G, Roach P. Cadmium, mercury and

selenium concentrations in mink (Mustela vison) from Yukon,

Canada. Sci Total Environ 2005b;351–352:523–9 [this issue].

Gerhart KL, White RG, Cameron RD, Russell DE. Body composi-

tion and nutrient reserves of Arctic caribou. Can J Zool 1996;

74:136–46.

Giesy JP, Kannan K. Global distribution of perfluorooctane sulfo-

nate in wildlife. Environ Sci Technol 2001;35:1339–42.

Giesy JP, Verbrugge DA, Othout RA, Bowerman WW, Mora MA,

Jones PD, et al. Contaminants in fishes from Great lakes—

influenced sections and above dams of three Michigan rivers

II Implications for health of mink. Arch Environ Contam Tox-

icol 1994;27:213–23.

Glooschenko V, Downes C, Frank R, Braun H, Addison E, Hickie J.

Cadmium levels in Ontario moose and deer in relation to soil

sensitivity to acid precipitation. Sci Total Environ 1988;

71:173–86.

Hayes RD. Numerical and functional responses of wolves, and

regulation of moose in the Yukon. M. Sc. Thesis, Simon Fraser

University, Vancouver, BC, Canada; 1995. 132 pp.

Health Canada. Food and drug regulations; 1991. 687 pp.

Heath AG. Water pollution and fish physiology. Boca Raton, FL7

CRC Press; 1987. 245 pp.

Hebert CE, Gamberg M, Elkin BT, Simon M, Norstrom RJ. Poly-

chlorinated dibenzodioxins, dibenzofurans and non-ortho sub-

stituted polychlorinated biphenyls in caribou (Rangifer

tarandus) from the Canadian Arctic. Sci Total Environ 1996;

185:195–204.

Hoekstra PF, Braune BM, O’Hara TM, Elkin B, Solomon KR, Muir

DCG. Organochlorine contaminants (OCs) and stable carbon

and nitrogen isotope profiles in arctic fox (Alopex lagopus) from

the Alaskan and western Canadian Arctic. Environ Pollut

2003a;122:423–33.

Hoekstra PF, Braune BM, Elkin B, Armstrong FAJ, Muir DCG.

Concentrations of selected essential and non-essential ele-

ments in arctic fox (Alopex lagopus) and wolverines (Gulo

gulo) from the Canadian Arctic. Sci Total Environ 2003b;

309:81–92.

Hoekstra PF, Braune BM, Wong CS, Elkin B, Muir DCG. Profile of

chlorinated contaminants, including selected chiral compounds,

in wolverine (Gulo gulo) livers from the Canadian Arctic.

Chemosphere 2003c;53:551–60.

Holz A, Sharpe MA, Constable M, Wilson W. Removal of contami-

nants from distant early warning sites in Canada’s Arctic. Edmon-

ton, Alberta7 Environmental Protection Service; 1987. 111 pp.

Kelly BC, Gobas FAPC. An Arctic terrestrial food chain bioaccu-

mulation model for persistent organic pollutants. Environ Sci

Technol 2003;37:2966–74.

Kelly BC, Gobas FAPC, McLachlan MS. Intestinal absorption and

biomagnification of organic contaminants in fish, wildlife, and

humans. Environ Toxicol Chem 2004;23:2324–36.

M. Gamberg et al. / Science of the Total Environment 351–352 (2005) 148–164164

Kelsall JP. The migratory barren-ground caribou of Canada. Cana-

dian Wildlife Service, Monograph no. 3, Queen’s Printer,

Ottawa; 1968. 339pp.

Kennedy D. Metals and organic contaminants in beaver and muskrat

in the Slave river Delta Area, NWT. In: Kalhok S, editor.

Synopsis of research conducted under the 1998–99 northern

contaminants program. Ottawa7 Indian and Northern Affairs

Canada; 1999. p. 127–32.

Kjellstrom T. Critical organs, critical concentrations and whole body

dose–response relationships. In: Friberg C, Elinder G, Kjell-

strom T, Nordberg GFCadmium and health: a toxicological

and epidemiological appraisal. Boca Raton, Florida7 CRC

Press; 1986. p. 231–46.

Lasat M. Phytoextraction of toxic metals: a review of biological

mechanisms. J Environ Qual 2002;31:109–20.

Li YF, Macdonald RW, Jantunen LMM, Harner T, Bidleman TF,

Strachan WMJ. The transport of h-hexachlorocyclohexane to

the western Arctic ocean: a contrast to a-HCH. Sci Total

Environ 2002;291:229–46.

Ma L, Komar K, Tu C, Zhang W, Cai Y, Kenelley E. A fern that

hyperaccumulates arsenic. Nature 2001;409:579.

Macdonald RW, Barrie LA, Bidleman TF, Diamond ML, Gregor

DJ, Semkin RG, et al. Contaminants in the Canadian Arctic: 5

years of progress in understanding sources, occurrence and

pathways. Sci Total Environ 2000;254:93–234.

Macdonald CR, Elkin BT, Roach P, Gamberg M, Palmer M. Inor-

ganic elements in caribou in the Yukon, NWT, and Nunavut

from 1992 to 2000: Spatial and temporal trends and the effect of

modifying factors. Unpublished manuscript prepared for the

Northern Contaminants Program, Ottawa, ON; 2002. 32 pp.

Martin JW, Smithwick MM, Braune BM, Hoekstra PF, Muir DCG,

Mabury SA. Identification of long-chain perfluorinated acids in

biota from the Canadian Arctic. Environ Sci Technol 2004;

38:373–80.

Muir D, Braune B, deMarch B, Norstrom R,Wagemann R, Gamberg

M, et al. Chapter 3: ecosystem uptake and effects. In: Shearer R,

editor. Canadian Arctic contaminants assessment report. Ottawa7

Indian and Northern Affairs Canada; 1997. p. 191–294.

Nilsson A, Huntington H. Arctic pollution. Arctic Monitoring and

Assessment Programme; 2002. 111 pp.

Nirwal AS. Environmental behaviour and fate of dichlorodiphenyl

trichloroethane (DDT) residues in a terrestrial Arctic ecosystem.

M.Sc. Thesis, Queen’s University, Kingston, ON; 2001.

O’Hara TM, Carroll G, Barboza P, Mueller K, Blake J, Woshner

V, et al. Mineral and heavy metal status as related to a

mortality event and poor recruitment in a moose population

in Alaska. J Wildl Dis 2001;37:509–22.

Outridge PM, MacDonald DD, Porter E, Cuthbert ID. An evalua-

tion of the ecological hazards associated with cadmium in the

Canadian environment. Environ Rev 1994;2:91–107.

Pasitschniak-Arts M, Lariviere S. Gulo gulo. Mamm Species

1995;499:1–10.

Poole K, Elkin B, Bethke R. Environmental contaminants in wild

mink in the northwest territories, Canada. Sci Total Environ

1995;160/161:473–86.

Puls R. Mineral levels in animal health: diagnostic data. Clearbrook,

BC7 Sherpa International; 1994. 356 pp.

Risenhoover K. Composition and quality of moose winter diets in

interior Alaska. J Wildl Manage 1989;53:568–77.

Scanlon P, Morris K, Clark A, Fimreite N, Lierhagen S. Cadmium

in moose tissues: comparison of data from Maine, USA and

from Telemark, Norway. Alces 1986;22:303–12.

Scheuhammer AM. The chronic toxicity of aluminium, cadmium,

mercury and lead in birds: a review. Environ Pollut 1987;

46:263–95.

Scheuhammer AM. Effects of acidification on the availability of

toxic metals and calcium to wild birds and mammals. Environ

Pollut 1991;71:329–75.

Smith TG, Armstrong FAJ. Mercury in seals, terrestrial carnivores,

and principal food items of the Inuit, from Holman, NWT. J Fish

Res Board Can 1975;32:795–801.

Snowshoe N. Uptake of contaminants in beaver and muskrat of

the Mackenzie River delta. Synopsis of research conducted

under the 2001–2003 northern contaminants program.

Ottawa7 Indian and Northern Affairs Canada 2004; 2003.

p. 336–41.

Suydam RS, Dickson DL, Fadely JB, Quakenbush LT. Population

declines of king and common eiders of the Beaufort Sea. Con-

dor 2000;102:219–22.

Thomas DJ, Tracey B, Marshall H, Norstrom RJ. Arctic terres-

trial ecosystem contamination. Sci Total Environ 1992;122:

135–64.

Thompson DR. Mercury in birds and terrestrial mammals. In: Beyer

GH, Heinz GH, Redmon-Norwood AW, editors. Environmental

contaminants in wildlife: interpreting tissue concentrations.

SETAC Spec Publ Ser. Geneva, Switzerland7 CRC Press Inc.;

1996. p. 341–56.

UNEP (United Nations Environment Program). Final act of the

conference of plenipotentiaries on the Stockholm convention

on persistent organic pollutants. Geneva, Switzerland7 UNEP;

2001. 44 pp.

Vandecasteele B, De Vosa B, Tack F. Cadmium and zinc uptake

by volunteer willow species and elder rooting in polluted

dredged sediment disposal sites. Sci Total Environ 2002;299:

191–205.

Wagemann R. Methylmercury and heavy metals in tissues of

narwhal, beluga and ringed seals. In: Kalhok S, editor. Synop-

sis research conducted under the 1997/98 northern contami-

nants program. Ottawa7 Indian and Northern Affairs Canada;

1999. p. 81–5.

Wania F, Haugen JE, Lei YD, Mackay D. Temperature dependence

of atmospheric concentrations of semi-volatile organic com-

pounds. Environ Sci Technol 1998;32:1013–21.

Wayland M, Gilchrist HG, Dickson DL, Bollinger T, James C,

Carreno R, et al. Trace elements in king eiders and common

eiders in the Canadian Arctic. Arch Environ Contam Toxicol

2001;41:491–500.

Weiner J, Krabbenhoft D, Heinz G, Scheuhammer A. Ecotoxicol-

ogy of mercury. In: Hoffman D, Rattner B, Burton G, Cairns J,

editors. Handbook of ecotoxicology, 2nd edition. Boca Raton,

Florida7 CRC Press; 2002.